Epidemiologic research has provided evidence for a relationship between development of musculoskeletal disorders and physical workplace factors such as repetitive work tasks, static contractions, and tiring postures (14). Neck/shoulder pain is especially widespread among office workers with intensive computer use (9,11), and the prevalence has been escalating through the past decades (6). During the same period, physical activity in general decreased (3,18).
It is unlikely that occupational use of computers will decrease in the future; thus, strategies for prevention for those without symptoms as well as rehabilitation for those with pain are pertinent. Regular physical activity can improve a number of chronic diseases (16), and in the past years, there has been a great deal of attention to motivate a more physically active lifestyle. Randomized controlled intervention studies have found positive effects on neck/shoulder pain regarding specific neck/shoulder muscle exercises (4,23,26), whereas exercise interventions without such specificity failed to reduce such pain conditions (20,22). Together, there is moderate evidence that intervention with specific neck/shoulder exercises can be beneficial upon treatment of chronic or frequent neck pain (24). However, there is a lack of studies concerning the effect of exercise intervention on prevention of future musculoskeletal disorders in those without symptoms. Also, there is inconclusive evidence on the influence of a high physical capacity to prevent development of neck/shoulder pain (7).
An often overlooked factor that may influence the outcome of intervention studies is the level of training attendance or "compliance", and it is yet to be determined which type of health-promoting workplace intervention that results in the highest compliance. Furthermore, it is unknown whether compliance per se or increased physical capacity as a result of training is decisive for reduction of musculoskeletal disorders.
To investigate this, we performed a randomized controlled trial over a 1-yr period in office workers where the interventions at the workplace were 1) specific resistance training (SRT) for the neck/shoulder muscles, 2) all-round physical exercise (APE) on a group level, and 3) reference intervention (REF) with counseling. We hypothesized that SRT would result in the highest compliance and gain in maximal shoulder muscle strength, and that this would lead to the best long-term reduction and prevention of neck/shoulder pain.
A randomized controlled intervention trial was performed in Denmark. The participants were office workers recruited from 12 geographically different located units of a national Danish public administration authority in the eastern part of Denmark. The flow of participants through the intervention is shown in Figure 1. Of the 2163 employees that were invited for the study, 1397 replied to the invitation, and 841 were willing to participate. Of these, 225 were excluded because of health risks or because too few at their unit wanted to participate, and therefore, 3 of the 12 units were not offered any intervention. The health risks were determined by the baseline questionnaire combined with an interview where blood pressure was measured, and it included conditions such as hypertension, disc prolapse, severe spinal disorders, history of severe trauma, or other factors such as pregnancy. In total, 616 participants (397 females aged 44.6 yr [weight, 68.2 kg; height, 1.68 m] and 219 males aged 45.7 yr [weight, 83.1 kg; height, 1.81 m]) participated in the baseline test of which further 24 were excluded because of the above criteria and 43 withdrew, leaving a total of 549 participants for the randomization. After the randomization, participants were excluded if they attracted a condition of the above exclusion criteria or if they were on maternity leave or other long-term absences leave (men and women). Five hundred and five participated in the midtest and 440 in the posttest. After the baseline measurements, the participants were randomized at the cluster level into the three intervention groups: 1) SRT (n = 180), 2) APE (n= 187), and 3) REF with counseling (n = 182). Clusters of participants were located on the same floor or in the same building and, thus, participated in the same intervention group. This was done to avoid contamination of the intervention and to enhance the compliance within the intervention groups, as well as to implement the study in a natural working environment. A total of 79 clusters were identified, and cluster size ranged from 1 to 25 participants; only nine clusters contained a single participant. A statistician at the National Research Centre for the Working Environment (Copenhagen, Denmark), specified the theoretical criteria for the randomization, and authors EAH and GS practically implemented those at the work sites. This cluster randomization resulted in three intervention groups that were comparable with regard to age, height, body mass, sex distribution, and intensity of neck/shoulder pain. The crew of physiologists and physiotherapists testing the participants was blinded with regard to which intervention group each participant belonged to. The data were analyzed according to the principle of "intention to treat."
The study protocol was approved by the local ethics committee (KF 01-201/04) and qualified for registration in the International Standard Randomised Controlled Trial Number Register on http://isrctn.org, and it has been assigned a unique trial identification number: ISRCTN31187106. All participants were informed about the purpose and content of the project, and gave written informed consent to participate.
Description of Interventions
All three intervention groups were allowed to spend a total of 1 h·wk−1 during working time for intervention activities.
Specific resistance training.
SRT was performed for the neck and shoulder muscles using a combination of traditional dynamic strengthening exercises with dumbbells for the muscles of the shoulder girdle and static exercises for the musculature of the cervical spine. The following dynamic exercises were performed for 2-3 sets of 10-15 repetitions: shoulder flexion ("front raise"), shoulder abduction ("lateral raise"), shoulder abduction with the thumbs facing downwards for increased activity of musculus supraspinatus (21), and shoulder elevation ("shrugs"). The first three exercises were performed in a range of motion from the neutral starting position to approximately 90° flexion/abduction, and the last exercise was performed in a full range of motion. The participants were encouraged to add weight when they were able to perform more than 15 repetitions. Repetition maximum (RM) test was performed only during the last month of the intervention period, where the participants performed as many repetitions as possible with the respective training weight of each exercise. The static neck exercises were performed in a sitting position with the cervical spine in the anatomical neutral position. An inelastic strap was positioned around the head of the participants and connected to a scale that was either handheld or secured to a hook fixed to the wall. The participants did not perform MVC to determine maximal muscle strength but were instructed to perform "near maximal contractions" of neck flexion, extension, and lateral flexion for repetitions of 5-s duration (25); according to a pilot study in our laboratory on a separate group of subjects, such instruction results in contraction intensities of approximately 70-80% MVC. The training sessions were ended with a high-speed dynamic power exercise, performed as 15-s all-out ergometer rowing (Concept2, Inc., Morrisville, VT) or kayaking (Dansprint, Vanløse, Denmark). Training was performed three times a week, and each session lasted 20 min. Two of the three weekly sessions were supervised by experienced instructors.
The training activity (type of exercise, load, sets, and repetitions) was registered in a diary during each session in this intervention group. All training was performed at the workplace during working hours.
All-round physical exercise.
The participants in this group were motivated to increase their level of physical activity during both leisure and at work, e.g., by taking the stairs instead of the elevator. Experienced instructors introduced different forms of physical exercise activities. The primary goal was to inspire the participant to integrate physical activity into their daily lifestyle in a motivating fashion. As part of the motivation, the participant filled in a "contract" in which they stated the ways in which they would include more physical activity into their daily lives, e.g., riding the bicycle for work instead of driving the car or joining the local fitness centre. Besides the introductory sessions, the frequency of visit from instructors ranged from one to four times a month. The physical activity at the workplace consisted of various forms of activities, e.g., steppers placed near the copying machines, punch bags in the hall, group sessions of Nordic walking, and exercise programs for both strength and aerobic fitness. At leisure time, the participants were encouraged to increase their physical activity levels by two campaigns: 1) ride the bike to work and 2) increase daily level of activity. The participants were provided with information on location and opening hours of local fitness clubs, swimming baths, etc.
The participants in this group were encouraged to form groups that should try to improve health and working conditions, e.g., through improved workplace ergonomics, stress management, organization of work, and cafeteria food quality. The participants themselves were responsible for organizing presentations about health-promoting activities that they found interesting, e.g., about diet, stress management, weight loss, meditation, relaxation, and indoor climate. Staff from our departments supported their work by helping to organize the presentations. The participants in this group received an equal amount of attention compared with the participants in the two other groups.
Maximal Voluntary Muscle Strength
Testing of maximal voluntary isometric muscle strength was performed according to a standardized procedure (5). For shoulder elevation and abduction strength, the participant was sitting upright in a height-adjustable chair, and two Bofors dynamometers were placed bilaterally 1 cm medial to the lateral edge of the acromion (10,19) and 1 cm proximal from the olecranon of the elbow joints (2), respectively. The participant was instructed to gradually build up the force over 5 s, then to keep the maximal force for about 2 s, and, finally, to lower the force slowly to zero. The MVC were performed at least three times for each exercise. If the third recording was more than 5% higher than the previous two recordings, a fourth test was performed, and a maximum number of five tests were performed. Strong verbal encouragement was given during all trials. There was no statistically significant difference in the change between left and right shoulder strength with the intervention; therefore, the average values of the left and right shoulders are reported.
The participants replied to an internet-based questionnaire with regard to neck and shoulder symptoms (at pre-, mid-, and postintervention) and participation in the intervention (at mid- and postintervention). Intensity of pain during the last 3 months was rated on a scale ranging from 0 to 9, where the following question was answered: "On average, how intense was your pain in [body part] during the last 3 months on a 0-9 scale?" (where 0 corresponded to "no complaints" and 9 to "pain as bad as it could be"). Duration of pain during the last 3 months was rated on a five-step ordinal scale: "How many days have you had trouble in [body part] during the last 3 months?" (0d, 1-7d, 8-30 d, >30 d, everyday).
Participation in the intervention was rated on a six-step scale: 1) "regular, each week at least 40 min," 2) "regular, each week between 20 and 40 min," 3) "irregular, each month at least 120 min," 4) "irregular, each month between 80 and 120 min," 5) "very irregular and less than 80 min per month," 6) "no participation at all." For the statistical analyses, these were collapsed to three groups: 1) regular participation (1 + 2 above), 2) irregular participation (3 + 4 above), and 3) seldom or no participation (5 + 6 above).
Cases and Controls
After the study was completed, the participants were divided into cases and controls based on the above questionnaire. This allowed us to investigate the effect of the interventions on reduction and prevention of symptoms in those with and without pain at baseline, respectively.
Neck cases and neck controls were identified as those reporting intensity of neck pain ≥3 and between 0 and 1, respectively, at a scale of 0-9 and duration of pain more and less than 7 d, respectively, during the last 3 months. The same criteria with regard to shoulder pain were used for identification of shoulder cases and shoulder controls. It should be noted that 64% of the neck cases also were shoulder cases, and 87% of the shoulder cases also were neck cases. The baseline characteristics of cases and controls are shown in Table 1.
All statistical analyses were performed using the SAS statistical software for Windows. Changes in main parameters were evaluated with mixed models. Variables included in the model were intervention group (SRT, APE, and REF) and time (pre, mid, post), as well as group by time. There were no gender by time effect (i.e., the change over time was not statistically significantly different between males and females); thus, all analyses were performed for males and females together. Multiple model regression analyses with backward elimination were performed to determine the relationship between changes in pain over time and the main variables (i.e., level of pain at pretraining, compliance, and change in muscle strength during the intervention). Frequencies (e.g., participation) were analyzed using chi-square test. An alpha level of 0.05 was accepted as significant. All values are reported as mean ± SE.
Regular participation for the participants who replied to the questionnaire (percentage of replies are given in parentheses) was attained in 54% (78%), 31% (81%), and 16% (80%) of the participants in SRT, APE, and REF, respectively, during the first half of the intervention period (SRT > APE > REF) (P < 0.0001) (Fig. 2A). During the second half, compliance generally decreased, and regular participation was achieved only by 35% (59%), 28% (71%), and 9% (70%) of the participants in SRT, APE, and REF, respectively (SRT and APE > REF) (P < 0.0001) (Fig. 2B).
In SRT, the number of training sessions was recorded in a diary and was 1.7 ± 0.1, 1.0 ± 0.1, and 0.2 ± 0.1 times per week for the participants who answered to the questionnaire that they participated regularly, irregularly, and seldom, respectively. The corresponding numbers during the second half of the intervention were 1.9 ± 0.1, 0.9 ± 0.1, and 0.2 ± 0.0 times per week, respectively. Compliance according to the questionnaire correlated with the participation based on training diary registrations (r = 0.72, P < 0.001).
Maximal Muscle Strength
The number of participants who completed all three test rounds was 48%, 45%, and 48% of the initial number of participants in SRT, APE, and REF, respectively.
For shoulder elevation strength, there was a significant group by time effect (P < 0.01). Post hoc tests showed that shoulder elevation strength was increased 11% (females, 12%; males, 10%) from 516 ± 19 to 573 ± 19 N pre- to posttraining in SRT (P < 0.0001) and 9% (females, 9%; males, 10%) from 586 ± 25 to 641 ± 26 N in APE (P < 0.0001), whereas no significant change occurred in REF (from 584 ± 23 to 605 ± 22, NS) (delta values in Fig. 3A). Furthermore, there was a significant increase from pre- to midintervention in SRT from 516 ± 19 to 544 ± 18 N (P < 0.05). There was no significant difference in these changes between genders.
For shoulder abduction strength, there was an overall increase from pre- to midtraining but no group by time effect. SRT increased 12% (females 11%, males 13%) from 246 ± 10 to 275 ± 12 N (P < 0.0001), APE 10% (females 9%, males 11%) from 289 ± 16 to 319 ± 18 N (P < 0.01), and REF 8% (females 11%, males 6%) from 270 ± 12 to 293 ± 13 N (P < 0.05) (delta values in Fig. 3B). There was no significant difference in these changes between genders. At the posttraining test, shoulder abduction strength was not significantly different from the pretraining value in APE and REF, whereas a tendency was observed in SRT (P= 0.06). Regression analyses showed that changes in muscle strength in response to the intervention were statistically unrelated to compliance, pain at baseline, and muscle strength at baseline. It should be noted that there was no statistically significant difference in the results between the left and right shoulders.
Neck Pain, Intensity, and Duration
In participants with neck pain at baseline, there was a significant group by time effect for intensity of pain (P < 0.05). The two groups that performed physical training decreased intensity of neck pain during the first half of the intervention period, SRT from 5.0 ± 0.2 to 3.4 ± 0.2 (P < 0.0001) and APE from 5.0 ± 0.2 to 3.6 ± 0.2 (P < 0.001), whereas no change occurred in REF (Fig. 4A). Pain intensity remained unchanged during the second half of the intervention period. Duration of pain during the last 3 months reported at baseline showed an overall decrease during the first half of the intervention period, SRT from 45 ± 2.9 to 25 ± 3.3 d (P < 0.0001), APE from 47 ± 2.9 to 26 ± 3.2 d (P < 0.001), and REF from 43 ± 2.8 to 30 ± 3.1 d (P < 0.0001), but there were no statistical significant difference between the three intervention groups. Regression analyses showed that changes in intensity of neck pain (in SRT and APE) were significantly related to pain intensity at baseline (participants with higher pain levels showed greater decrease in pain) and changes in shoulder elevation strength; however, the explained variance was low (R2 = 0.11, P < 0.05). Compliance was unrelated to changes in neck pain.
In neck controls, there was a general increase over time in pain intensity (P < 0.0001), but there was no statistical significant difference between the groups (Fig. 4A). Duration of pain remained unchanged in all three groups.
Shoulder Pain, Intensity, and Duration
In participants with shoulder pain at baseline, there was a decrease over time from pre- to midtraining in pain intensity and duration in both the right and left shoulders, but there was no statistically significant difference in the change between groups (Fig. 4B). Regression analyses showed that neither compliance nor changes in muscle strength could explain the decrease in shoulder pain. However, participants with higher pain levels at baseline showed greater decrease in pain, although the explained variance was low (13% and 11% for right and left shoulders, respectively, P < 0.05).
In shoulder controls, there was a group by time effect for both intensity and duration of pain for the right shoulder (P < 0.01). Intensity and duration of pain increased more in REF than in SRT and APE (P < 0.01) (Fig. 4B). For the left shoulder, there was also a general increase in duration and intensity of pain, but no statistical significant difference between the groups. A significantly greater proportion of the shoulder controls in REF compared with SRT and APE were defined as shoulder cases after the intervention (19%, 5%, and 4% in REF, SRT, and APE, respectively, P < 0.05).
Training Load and RM Test in SRT
Training load in SRT increased approximately twofold during the 1-yr intervention period (shown for one of the exercises, "shrugs", in Fig. 5).
The number of repetitions that could be performed to failure with the respective training loads of 18.3 ± 0.9 and 4.9 ± 0.3 kg during the last month of training was 34.1 ± 2.1 repetitions and 20.6 ± 1.8 repetitions for the shoulder elevation ("shrugs") and shoulder abduction exercises ("lateral raise"), respectively.
The present study showed that in shoulder controls, development of shoulder pain over a 1-yr period was significantly less in the two exercise groups compared with REF. Furthermore, approximately one-fifth of the original shoulder controls in REF could be classified according to the more severe definition of "shoulder cases" after 1 yr, whereas this was the case in only a few percentages of the participants in the two exercise groups. Thus, the present findings suggest that both types of exercise interventions used in the present study can prevent development of shoulder pain. Prevention of shoulder pain development by exercise may help office workers and workers in the industries who are currently pain free and industries by saving compensation cost over the long term.
The present study supports the finding that specific exercises can reduce intensity of neck pain in cases (4,23,26) and suggests that increasing general physical activity can also have a beneficial effect. Previous studies with physical exercise intervention without specific neck exercises did not find a positive effect on neck pain (20,22). The difference between those studies and the present may be that APE obviously included exercises that strengthened the muscles of the neck and shoulder as seen from the 9% increase in shoulder elevation strength in this group. In SRT, the gains in shoulder elevation strength appeared to be modest compared with gains in neck/shoulder muscle strength in previous studies of similar or shorter duration (1,26). These results may be explained by a combination of a relatively low compliance and a lower than intended training intensity. Gains in muscle strength are typically found in response to resistance exercises involving high intensity (12) and may be an indirect measure or a proxy of specific neck/shoulder muscle training compliance. Overall, the results of the two exercise groups indicate that physical activities leading to gains in local muscle strength may be helpful in reducing neck pain symptoms. The REF intervention had low compliance and limited impact on neck/shoulder pain, which is in agreement with previous intervention studies on counseling and work health promotion (8,17).
The highest level of participation was found among the participants who performed SRT at the workplace; however, overall compliance decreased during the second half of the intervention period. Thus, strategies to keep up long-term motivation of the employees to exercise should be developed.
In the present study, we were not able to show a significant correlation between compliance and reduction in pain, nor between compliance and gains in strength. However, there was a weak but significant correlation between reduction in neck pain and gains in muscle strength. A recent study showed an association between reduction in neck pain symptoms and amount of training based on training diaries (15). The present finding implies that participation in physical activities per se cannot explain the reduction in neck pain, but rather, the intensity and total amount of work performed during training appears to be important (15). In the present study, pain and compliance were assessed by questionnaires at three static time points and may, thus, have been biased by recall inaccuracy (13). On the other hand, in the present study, there was a clear association between compliance based on the questionnaire and the actual number of training sessions based on training diary registrations in SRT. Although training diaries were only kept in SRT, this finding indicates that questionnaire-based compliance is representative of actual participation in the intervention.
The change in shoulder abduction strength may be related to seasonal variation, in that all three groups increased from pre- to midintervention, and returned to levels that were not statistically different from baseline at postintervention (Fig. 3B); e.g., a higher level of leisure activity such as gardening and outdoor physical activities in the spring and summer may explain these results.
Adverse events (e.g., shoulder ache, neck ache, or backache due to overexertion or incorrect strength training technique) were minor and transient probably because of gradual progression in training load and supervision of the training by experienced instructors taking immediate action in case of minor complaints.
In conclusion, SRT and APE resulted in clinically relevant reductions of neck pain in those with symptoms and prevention of shoulder pain in those without symptoms, although only minor gains in muscle strength were found. Compliance was highest in SRT but generally decreased over time. Thus, a challenge for future studies may be to enhance long-term compliance.
Warm thanks to technician Dorte Ekner and physiotherapist Klaus Hansen for outstanding technical support. The study was financially supported by funding from the Ministry of Culture Committee on Sports Research N200310016 and the National Board of Health under the Ministry of the Interior and Health. Training was supervised by Dansk Firmaidrætsforbund (DFIF) regarding the all-round physical exercise.
1. Ahlgren C, Waling K, Kadi F, Djupsjobacka M, Thornell LE, Sundelin G. Effects on physical performance and pain from three dynamic training programs for women with work-related trapezius myalgia. J Rehabil Med
2. Backman E, Johansson V, Hager B, Sjoblom P, Henriksson KG. Isometric muscle strength
and muscular endurance in normal persons aged between 17 and 70 years. Scand J Rehabil Med
3. Booth FW, Chakravarthy MV, Gordon SE, Spangenburg EE. Waging war on physical inactivity: using modern molecular ammunition against an ancient enemy. J Appl Physiol
4. Chiu TT, Lam TH, Hedley AJ. A randomized controlled trial on the efficacy of exercise for patients with chronic neck pain. Spine
5. Essendrop M, Schibye B, Hansen K. Reliability of isometric muscle strength
tests for the trunk, hands and shoulders. Int J Ind Ergon
6. Hakala P, Rimpela A, Salminen JJ, Virtanen SM, Rimpela M. Back, neck, and shoulder pain in Finnish adolescents: national cross sectional surveys. BMJ
7. Hamberg-van Reenen HH, Ariens GA, Blatter BM, van Mechelen W, Bongers PM. A systematic review of the relation between physical capacity and future low back and neck/shoulder pain. Pain
8. Hillsdon M, Thorogood M, White I, Foster C. Advising people to take more exercise is ineffective: a randomized controlled trial of physical activity
promotion in primary care. Int J Epidemiol
9. Jensen C. Development of neck and hand-wrist symptoms in relation to duration of computer use at work. Scand J Work Environ Health
10. Jensen C, Vasseljen O, Westgaard RH. The influence of electrode position on bipolar surface electromyogram recordings of the upper trapezius muscle. Eur J Appl Physiol Occup Physiol
11. Juul-Kristensen B, Sogaard K, Stroyer J, Jensen C. Computer users' risk factors for developing shoulder, elbow and back symptoms. Scand J Work Environ Health
12. Kraemer WJ, Adams K, Cafarelli E, et al. American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Med Sci Sports Exerc
13. Miranda H, Gold JE, Gore R, Punnett L. Recall of prior musculoskeletal pain. Scand J Work Environ Health
14. National Research Council and Institute of Medicine. Musculoskeletal Disorders
and the workplace-Low Back and Upper Extremities. Washington (DC): National Research Council, Institute of Medicine; ISBN-10: 0-309-07284-0; 2001.
15. Nikander R, Malkia E, Parkkari J, Heinonen A, Starck H, Ylinen J. Dose-response relationship of specific training to reduce chronic neck pain and disability. Med Sci Sports Exerc
16. Pedersen BK, Saltin B. Evidence for prescribing exercise as therapy in chronic disease. Scand J Med Sci Sports
. 2006;16(Suppl 1):3-63.
17. Proper KI, Hildebrandt VH, Van der Beek AJ, Twisk JW, Van Mechelen W. Effect of individual counseling on physical activity
fitness and health: a randomized controlled trial in a workplace setting. Am J Prev Med
18. Sjol A, Thomsen KK, Schroll M, Andersen LB. Secular trends in acute myocardial infarction in relation to physical activity
in the general Danish population. Scand J Med Sci Sports
19. Sogaard K, Fallentin N, Nielsen J. Work load during floor cleaning. The effect of cleaning methods and work technique. Eur J Appl Physiol Occup Physiol
20. Takala EP, Viikari-Juntura E, Tynkkynen EM. Does group gymnastics at the workplace help in neck pain? A controlled study. Scand J Rehabil Med
21. Townsend H, Jobe FW, Pink M, Perry J. Electromyographic analysis of the glenohumeral muscles during a baseball rehabilitation program. Am J Sports Med
22. Viljanen M, Malmivaara A, Uitti J, Rinne M, Palmroos P, Laippala P. Effectiveness of dynamic muscle training, relaxation training, or ordinary activity for chronic neck pain: randomised controlled trial. BMJ
23. Waling K, Sundelin G, Ahlgren C, Jarvholm B. Perceived pain before and after three exercise programs-a controlled clinical trial of women with work-related trapezius myalgia. Pain
24. Ylinen J. Physical exercises and functional rehabilitation for the management of chronic neck pain. Eura Medicophys
25. Ylinen J, Ruuska J. Clinical use of neck isometric strength measurement in rehabilitation. Arch Phys Med Rehabil
26. Ylinen J, Takala EP, Nykanen M. Active neck muscle training in the treatment of chronic neck pain in women: a randomized controlled trial. JAMA